Abstract
In this paper, the self-adaptive superpixels are generated based on a neural network model. Superpixels are clusters of pixels, which can simplify the expression of images. Superpixels are widely used in the field of video/image processing. However, existing algorithms are mainly based on hand-crafted features, which will lose the details of the images. We use the neural network model to extract the deep features of the pixels instead of the hand-crafted features. A predicted object area is obtained according to the results of the neural network models. Self-adaptive superpixels are generated by the clustering algorithm based on the deep features of the pixels and the predicted object area. Finer superpixels are generated in the object area, and coarser superpixels are generated in background area. The generated self-adaptive superpixels can represent the image in a concise way and improve the segmentation accuracy. Experimental results show that the proposed algorithm outperforms several state-of-the-art methods on the BSDS500 dataset.
Highlights
The concept of superpixels was first proposed by Ren [1] in 2003
We propose a self-adaptive superpixel algorithm, which is based on deep learning and clustering
Model features refer to the features of each pixel predicted by the deep neural network model and use Principal Component Analysis (PCA) to reduce the dimension
Summary
The concept of superpixels was first proposed by Ren [1] in 2003. In the field of image segmentation, superpixels have the same meaning as image over-segmentation. The principle of adaptability is to generate more superpixels in the object area and less superpixels in the background area. The object area is relatively more important than the background area It can adaptively assign more superpixels in the object area. Self-adaptive superpixels can better express the image and retain the detailed features of the image. The main contributions of our paper are as follows: 1) We use different network models to extract different deep features to generate superpixels, such as image classification network, image semantic segmentation network, and saliency object detection network. 2) We can obtain the adaptability of superpixels based on the result of the network model. Our method can adaptively assign more superpixels in the object area, which can better express the image and retain the detailed features of the image. Our method can adaptively assign more superpixels in the object area, which can better express the image and retain the detailed features of the image. 3) Our method performs better than the state-of-the-art approaches, which helps to improve the application of image processing based on superpixels
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
